A structural link to a prior business caps this at Established — a reinstated shell never reads as new.
Decision-makerIs this the right person? The founder, not the filer.
0–100tune it
✓Named on the state filing
✓Their own registered agent
✓Address matches the business
✓Name appears in the business name
✓How many businesses share this person
✓How common the name is
✓Professional filers flagged and set aside
Decision makerLikelyProbableUncertainUnlikely
Tiers and the families of signals are on every record. How they combine is ours.
Contact ConfidenceIs it really them? Email and phone matched to the actual person.
0–100tune it
✓Email contains their own name
✓Deliverable, independently verified
✓Mobile phone tied to their filing
✓Residential address raises confidence
✓Self-represented on the filing
✓Their name confirmed across sources
✓How strong the original match was
VerifiedLikelyPossibleUncertain
Cross-checked against deliverability and how the contact was matched. How it combines is ours.
The nightly buildwhat happens to one filing
1
Pull the filings
Every new business the state published that day, as the state published it.
State filings · 6 states
2
Screen out what isn't a business
Mail drops and formation suites, holding shells, single-property LLCs. 83% pass.
Address intelligenceRegistered-agent intelligence
3
Find the decision-maker
The officer or owner named on the filing, not the attorney or service who filed it. Every U.S. surname (162,249) and given name (6,782) with how common each is, so a match on a rare name counts for more than a match on a common one.
US Census surnamesSSA given namesName intelligenceProfessional-filer detection
4
Place the business
Home or commercial address, neighborhood and its income, metro, time zone.
US CensusProperty recordsUSPS · neighborhoods · time zonesAddress + property intelligence
5
Name the line of business
Industry, NAICS and MCC from the business name and filing; AI reads the hard ones.
NAICS · SIC · MCC taxonomiesIndustry classification
6
Find and verify the channels
Mobile or landline, in service or not; email deliverable, and active in the last months.
Phone recordsEmail verificationContact intelligence picks the route
7
Pick the best person, phone and email
Among everyone named on the filing and every channel found, one primary of each.
Primary contact · phone · email selection
8
Score
Can we reach them, and is this the right person. Then the record is yours.
ReachabilityDecision-maker
Public recordVerification partnerOurs
Run by AI agents, reviewed by AI agents
An agent team runs the build nightly, briefs on each state's data every morning, and reviews every change to the intelligence before it ships. The build gets better every week without waiting on us.
Sources and steps as of September 2026, from the platform's source registry; the count is checked by the build every time the code changes.
The bank is chosen in the first few weeks.
Every new business needs an operating account before it trades. We have sold into this moment before: the case study below is our own, from another first-week purchase.
The moment that matters. A new owner has no bank, no card and no payroll yet. Each is a first decision, made before they appear in the databases your competitors use.
Files, gets an EIN
days 0–14
Needs an account to keep business and personal money apart
The state publishes the filing; the tax ID follows within days. GoodLeads reaches the owner here.
Chooses a bank
days 15–45
Opens an account and decides how to take payments
The account, the card and the way to get paid are picked together. Whoever reaches them first makes the case first.
Shows up everywhere
days 60–90
Appears in mainstream business databases
Most data vendors arrive here, after the account is open. Timeline is indicative.
Every record carries days since filing, so each stage can get its own message. What to send at each stage is on the Formation Watch page.
GO
Gib Olander · Founder, GoodLeadsFormerly Chief Product Officer, a national business-formation company
The moment
Over three-plus years, at least 670,000 businesses saw an offer in our checkout, at the moment they formed, before they had a name, a customer or a dollar of revenue.
What we did
Launched a business phone line into that moment at $9 a month: zero brand awareness, very little functionality. The timing was the pitch.
What happened
Revenue tripled in four years on formation-moment upsell. The phone line became a $15M business in three years, and is on its way to $18M this year.
"They didn't need to know who we were. They needed it that week. Nobody else was there."
1 in 4
said yes with nothing going for the offer but the moment
25% attach at checkout, no brand, few features. That is the attach rate of a need, not a pitch.
91%
are starting their first business
No incumbent bank relationship for the business to unwind. ZenBusiness, State of Entrepreneurship Report 2026.
65%
filed from a home address
The founder is a household you can reach, and often one you already bank. GoodLeads analysis, August 2026 filings.
~70%
self-funded from personal savings
The first deposit comes out of a personal account. ZenBusiness, State of Entrepreneurship Report 2026.
Will a days-old business respond?They bought from a brand they had never heard of, before revenue, before doors opened. The need was the gate, not the brand.
That was a checkout. We'd be reaching out.Right. That company owned the moment because it sat in the flow. The feed puts you in the same moment within days of the filing; the pilot tunes the outreach.
Is there credit to write on day one?Open with deposits and payments; credit follows. A brand-new business has no trading history, so four of the five personas start with an account, a card and a way to get paid.
Why can't we get this elsewhere?Business databases find a new company 60 to 90 days in, after it has an account, a card and a listing. We deliver the founder the morning after the state posts the filing.
How may the data be used?Marketing and sales outreach, not underwriting. Every lending product runs through your normal application and credit process.
163,000new businesses a month in our six states: nearly 2M a year
3 in 10of U.S. likely-employer business applications start in those six states
Before the account existswe deliver the founder. Business databases pick up the company 60–90 days later.
Every new business around your branch, sorted into five personas.
Each business lands in one persona: what the filing is decides first, then the trade, then where they work.
NearWithinFormed in the last
Loading the businesses…
0
new businesses
Click a pin to see the business · drag or scroll to look around
Show on map ·
Formation Volume · August 2026 · all six states
5,257new businesses per day
36,799per week
162,967in August
33%of U.S. likely-employer applications start in our six states
Each bar is one day. Business days carry the volume (median 6,761, peak 8,136 on Aug 10); weekends run about a quarter of that. 31 days, no missed days. Likely-employer share: U.S. Census Bureau, Business Formation Statistics, August 2026.
Top metros · August · operating businesses
Metro
/day
/week
Miami – Fort Lauderdale
578
4,046
New York City (NY side)
379
2,651
Dallas – Fort Worth
310
2,171
Houston
263
1,839
Tampa – St. Petersburg
224
1,569
Denver
203
1,421
Orlando
184
1,286
Washington DC (VA side)
127
891
Operating businesses only; holding and single-property entities are screened out first.
By business type · August · per week by state
Business type
/day
/week
FL
TX
NY
CO
VA
CT
Retail store
162
1,132
413
215
198
166
96
44
General contractor
83
584
193
165
91
61
46
28
Cleaning services
78
545
292
79
49
60
47
19
Marketing / advertising
75
528
217
109
73
71
43
16
Restaurant
74
519
180
103
114
54
44
24
Trucking
69
484
180
182
42
32
38
10
Auto repair
56
391
173
91
45
35
33
15
Salon / barber
53
368
174
64
50
36
32
12
Freight & logistics
48
338
140
119
27
21
24
9
Fitness
38
266
121
56
29
30
23
7
Dots show the persona each type falls into. Counts are floors: about 7 in 10 businesses carry a type today.
Real people, days into their business, with the proof on every fact.
One founder per persona from this market. We found the decision-maker, tied them to the business they just formed, and every fact names where it came from. Each card opens the full GoodLeads profile and how it was built.
91%are starting their first business65%filed from a home address~70%self-funded from personal savings~1 in 10may already bank with Your Bank
Portraits are AI-generated illustrations chosen by each business's industry. They are not photos of these founders.
First-business and self-funded shares: ZenBusiness, State of Entrepreneurship Report 2026. Home address: GoodLeads analysis of August 2026 filings in six states. Product sets are illustrative; each bank sets its own products, offers and eligibility, and every lending product runs through its normal application and credit process.
Each founder lights up a persona, and the products that persona needs.
← → flips founders · click a persona to jump to its founder
36 sources and engines behind every record. A filing scraper has one.
Built every night from the source: 18 data sources refined by 18 intelligence engines of our own. Every fact on a record names where it came from, and the scores tell your bankers who to call first.
18external data sources: 14 public feeds (seven states' filings, Census, property records) and 4 partners for phone, email, identity and AI reading
18intelligence engines of our own, run in order on every record
9reference corpora underneath: every U.S. surname and given name with its frequency, NAICS · SIC · MCC taxonomies, USPS geography, area codes, neighborhoods, time zones
Every nightthe day's filings become records before you wake up. The count is pinned to the code.
The nightly buildwhat happens to one filing
1
Pull the filings
Every new business the state published that day, as the state published it.
State filings · 6 states
2
Screen out what isn't a business
Mail drops and formation suites, holding shells, single-property LLCs. 83% pass.
Address intelligenceRegistered-agent intelligence
3
Find the decision-maker
The officer or owner named on the filing, not the attorney or service who filed it. Every U.S. surname (162,249) and given name (6,782) with how common each is, so a match on a rare name counts for more than a match on a common one.
US Census surnamesSSA given namesName intelligenceProfessional-filer detection
4
Place the business
Home or commercial address, neighborhood and its income, metro, time zone.
US CensusProperty recordsUSPS · neighborhoods · time zonesAddress + property intelligence
5
Name the line of business
Industry, NAICS and MCC from the business name and filing; AI reads the hard ones.
NAICS · SIC · MCC taxonomiesIndustry classification
6
Find and verify the channels
Mobile or landline, in service or not; email deliverable, and active in the last months.
Phone recordsEmail verificationContact intelligence picks the route
7
Pick the best person, phone and email
Among everyone named on the filing and every channel found, one primary of each.
Primary contact · phone · email selection
8
Score
Can we reach them, and is this the right person. Then the record is yours.
ReachabilityDecision-maker
Public recordVerification partnerOurs
ReachabilityCan we reach them? Every record, 0–100 and a tier.
✓Phone in service, mobile over landline
✓Email deliverable and active recently
✓Identity confirmed across sources
✓Address type: home or commercial
✓Area code fits the state
✓Line of business known
On FireVery HotHotWarmCold
Decision-makerIs this the right person? The founder, not the filer.
Tiers and the families of signals are on every record. How they combine is ours.
How these founders scored
Founder
Reachability
Decision-maker
Built for a bank's review
Data you can put in front of compliance.
✓
Sources · public state filings, with phone and email data licensed with marketing rights
✓
Owner, not the filer · registered agents, law offices and repeat filers are identified and set aside
✓
Provenance on every field · public record, verified by a partner, or our own intelligence, with a timestamp
✓
Clean before it ships · shells, mail drops and holding companies never reach you
✓
Scored three ways · New Business, Decision-maker and Contact Confidence
✓
Right channel · ad-platform audiences only on first-party matched customers, never uploaded from our data
How the data may be used
Marketing and outreach, not underwriting.
Licensed for marketing and sales outreach. Every lending product in this demo runs through your normal application and credit process. Formation Watch is first-party marketing to your own customers: the match runs inside your bank, and your list never leaves your environment.
The GoodLeads guarantee. If a record isn't accurate, we replace it, free. You never pay for data that doesn't hold up.
Run by AI agents, reviewed by AI agents
An agent team runs the build nightly, briefs on each state's data every morning, and reviews every change to the intelligence before it ships. The build gets better every week without waiting on us.
Sources and steps as of September 2026, from the platform's source registry; the count is checked by the build every time the code changes.
Formation Watch: your customer just started a business.
Founders flow into your systems; your customer list never comes to us. The match happens inside Your Bank, and every founder becomes a person you can reach across every channel you run.
1
The feed
Every day, the founders who filed. Each one arrives with the keys to reach them.
New founders · daily, by API or fileFrom this market · click a row
Founder
Formed
Mobile
Email
Address
Persona
The feed lands in your CDP or CRM. Nothing comes back to us.
2
The match
Inside Your Bank, our keys run against your customers. Any one of them is enough.
Inside Your Bank · your customer list never leaves this box
1 · The feed arrives
Today's founders, newest first, with the keys.
2 · Your customer database
Our keys run against your rows. Any one is enough; hashed keys reach the audience platforms too.
3 · What happens next
3
The reach
The identity goes out on every channel you run, in time order clockwise from 12. Each spoke shows the key that carries it and the message written from our record.
Show it as
Illustrative ads styled after each platform's native format. First-party audiences apply only to customers Your Bank already holds; founders who aren't customers are reached by mail, verified email, banker calls and branch invites under your consent rules. Customer rows are placeholders.
See your five personas, sized for your markets.
Tell us your branch and banker locations. We size each persona around them with live counts, walk through real founder records, and show how a Formation Watch match test works. One analyst, one afternoon, nothing sent to anyone.
Scroll to see the 120-day pilot that follows
Sixteen weeks, five readouts, the decision at the endConnects, accounts and bounces come back weekly. We bring the data side.
Wk 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Matchcatch-up file · your box
Match
Dial increative · offer · motion
A/B creative · offer · motion
Scalewinners only
Scale the winners · both audiences
Already yoursowned channels · holdout
App · online banking · email · audiences
Not yet yoursdirect channels · control
Email · mail · banker · branch
Every filedaily feed
Scorecard + vendor packet
Match rate+ response by day
Connects+ bounces
Winners+ first accounts
Cost per accountfirst read vs baseline
Decisiongo-live read · business case
A 120-day pilot that becomes your business case.
Your markets. Match first, dial in, scale the winners. The business case at the end, with the go-live read in hand.
How many are already ours?Match test · week 1
What creative, offer and motion work?Dial in · weeks 2–6
Does it lift accounts?Scale · weeks 7–16
Is the data clean and ready to run?Scorecard · every file
What does a new account cost?Business case · week 16
Sixteen weeks, five readouts, the decision at the endConnects, accounts and bounces come back weekly. We bring the data side.
Wk 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Matchcatch-up file · your box
Match
Dial increative · offer · motion
A/B creative · offer · motion
Scalewinners only
Scale the winners · both audiences
Already yoursowned channels · holdout
App · online banking · email · audiences
Not yet yoursdirect channels · control
Email · mail · banker · branch
Every filedaily feed
Scorecard + vendor packet
Match rate+ response by day
Connects+ bounces
Winners+ first accounts
Cost per accountfirst read vs baseline
Decisiongo-live read · business case
Week 1 · Match test
How many founders are already your customers?
What
Catch-up file: last 120 days, your markets
Keys: mobile · email · name + ZIP
Runs inside your systems; nothing comes back
Hashed on request
You get
Match rate and count
Split by persona and by day since formation
Which key matched
The response-by-days curve
Costs you
One analyst, one afternoon
No send, no consent, no exposure
~10%our assumption. Replaced by your number.
Gate → match rate sizes Already yours and the platform audiences.
Weeks 2–16 · Already yours
Your customer just started a business.
WhoMatched founders · catch-up cohort first, then monthlyChannelsApp · online banking · email · Customer Match · Custom AudiencesAgainstA holdout of matched foundersMeasuredPersonal customer → business account · products per account · days to first depositWhyThe first deposit comes out of an account you already hold.
Weeks 2–16 · Not yet yours
A founder in your market, choosing a bank.
WhoUnmatched founders · daily feed only, fresh recordsChannelsEmail · mail · banker call · branch inviteAgainstYour current new-business source, worked the same wayMeasuredConnect rate by persona and day · qualified conversations · new accounts · bouncesWhyNobody else has them yet. Days, not weeks.
Dial in, then scaleWeeks 2–6 test. Week 7 the winners run.
Creative
Congrats on {business}Business and personal, apartAccount, card reader, one place
Offer
Business checking, opened onlineChecking + card acceptanceOperating account + banker intro
Motion
App + emailEmail + banker callMail + branch
Winner by persona, by dayEach persona and each stage picks its own winner. Ours to write from the record; yours to run.
With every file: data-quality scorecard
Yours arrives with each delivery.
With every file: vendor packet
For procurement and your data team. The paperwork, already done.
✓
Provenance on every fact
✓
Public-record source · state filings
✓
Line type on every phone · mobile flagged for text and call
✓
Your suppression list applied first
✓
Hashed match on request
✓
Licensed for marketing and outreach · not underwriting
✓
Data agreement and security questionnaire, ready to sign
What we need from you
An owner for the match
A holdout and a control group
Your markets, 120 days
Weekly outcomes back: connects, accounts, bounces
Week 16: the business case
Cost per new account, next to your baseline, with the first cohort fully live.
Wins → a standing daily feed, the "new business owner" tag in your CRM, first access to every new state and persona, vendor review done.
Measured in your systems against your baseline. Lending products run through your normal application and credit process.